In new research, Cree Jones, Tyler B. Lindley, and Thomas Smith investigate how restrictions on the president to remove independent agency officials affect agency behavior. Such restrictions have historically had surprisingly little effect. However, recent political polarization has drastically increased the importance of removal restrictions in blocking political influence.
Among the many agencies that form our administrative bureaucracy, a significant number are “independent agencies,” which are led by government officials that Congress chose to protect from at-will removal by the president. The provisions shielding these agency leaders are commonly referred to as removal restrictions or protections. Now, as President Donald Trump asserts the authority to fire almost all independent agency leaders, the Supreme Court is considering whether removal restrictions violate the U.S. Constitution and the separation-of-powers doctrine.
If the Court agrees, the president’s critics argue, agencies will be ruled by political and partisan whims, as opposed to technical expertise and judgment. Supporters, if not of President Trump then the principle of ending removal restrictions, argue that agency independence results in an unelected, unaccountable, and constitutionally impermissible technocratic entity. The missing component to the debate is whether (and how much) statutory removal protections actually affect agency behavior.
In “Testing the Independence Hypothesis,” we tackle the empirical question. Our research shows that these restrictions were historically redundant with bipartisan political norms that accepted agency independence, and so had no practical effect on agency behavior. But recent political polarization has given removal restrictions real-world effect. Rulemaking, enforcement actions, and agency personnel are all potentially affected when the heads of these agencies are protected from the president’s control.
Modern stakes: the Slaughter case
Removal restrictions entered the news once again in March 2025 when President Trump sent a letter to Commissioner Rebecca Slaughter, one of the two Democratic commissioners of the Federal Trade Commission (FTC), purporting to remove her from her post. Federal statute provides that FTC commissioners—of which there are normally five, with up to three from the president’s own political party— may be removed by the President for inefficiency, neglect of duty, or malfeasance in office.” Despite this statutory restriction, President Trump refused to assert inefficiency, neglect, or malfeasance, instead resting on his “authority under Article II of the Constitution,” as the head of the executive branch, to remove officers who wield “substantial executive power.”
A district court and a divided panel of three appellate judges in D.C. held that the removal restriction was constitutional and thus that President Trump’s purported firing was unlawful. The district court ordered Slaughter’s reinstatement. Shortly after, the Supreme Court temporarily halted the reinstatement order and is now set to rule on the constitutional question.
The president’s constitutional removal power has been the subject of debate in Congress since as early as 1789, yet the question remains unsettled. In 1935, the Supreme Court held that the statute protecting FTC commissioners like Slaughter was constitutional and justified the ruling on the unique scope of the FTC’s power that did not include any constitutional executive power. In recent years, though, the Supreme Court has grown skeptical of Congress’s power to insulate executive officials from the president’s control. In 2020, the Supreme Court said that the 1935 Court was likely wrong about the scope of the FTC’s power at that time but that, in any event, the modern FTC is much different because it exercises vast executive power. In Slaughter’s case, the Court will have to determine whether those differences should yield a different outcome, or whether it should overturn the 1935 decision altogether. And, of course, in the background lurks the independence of the most influential (and arguably most popular) independent agency: the Federal Reserve Board.
The true impact of removal restrictions
To evaluate the impact the end to removal restrictions has on the actions of independent agencies, we studied how eliminating removal restrictions for the Public Company Accounting Oversight Board (PCAOB) changed leadership behavior. This intervention in the PCAOB’s structure, which occurred in 2010, represents the Supreme Court’s first modern move toward recognizing the president’s constitutional removal power.
The PCAOB is a government agency that oversees audits of American public corporations. The independent agency is embedded in the Securities and Exchange Commission, which chooses the five members of the PCAOB’s leadership board. The Supreme Court ruled that removal restrictions for the PCAOB members were unconstitutional, effectively eliminating removal protections for the PCAOB leadership. Because the PCAOB’s unique structure, embedded in the SEC, another independent agency, creates a dual layer of removal protections, the Court’s decision affected only the PCAOB without undermining removal protections at other independent agencies.
Using that Supreme Court decision as a natural experiment, we took the number of early departures by PCAOB leaders (before their terms expired) and compared that data with early departures of leaders at other independent agencies that have similar features but, importantly, continued to benefit from their removal protections beyond 2010.
This approach, known as a difference-in-differences design, estimates the causal relationship between removal protections and early departures of agency leaders by using early departures at other independent agencies to isolate the effect of the Court’s decision on the PCAOB from general trends. This statistical design provided an estimate of how the composition of PCAOB leadership would have changed had the removal protections remained in effect. Early departures are instructive as to agency behavior because they produce personnel changes that could dramatically alter the agency’s policy and enforcement decisions, particularly when policy disagreement is the primary driver of the changes. And even when the agency heads do not leave early, the mere threat of being fired might cause them to conform their behavior to the President’s will.
Our analysis demonstrates that for at least eight years following the elimination of removal restrictions at the PCAOB, there was no noticeable difference in the frequency of early departures compared with similar independent agencies. Even without the legal effect of statutory protections, PCAOB leaders were not any more likely to leave their posts early, suggesting that something other than removal restrictions was protecting agency leaders’ jobs.
Then, from 2018 to 2022, judicial elimination of removal restrictions caused more than a 300% increase in early departures at the PCAOB, which included the firing of all PCAOB members following the 2020 elections. The timeline covers the last two years of the first Trump administration and the first two years of the Biden administration. These results suggest that circumstances changed during this period: enforceable statutory removal restrictions continued to protect agency leaders at other agencies, while the PCAOB’s leaders became more vulnerable to early departure.
Our results shed light on the behavioral effect of statutory removal restrictions. Although removal restrictions historically might have done very little to change agency behavior, they have recently become a meaningful mechanism insulating independent agencies from the president’s influence. Invalidating them today would open agencies up to such influence.
Understanding our results
Why did removal protections have so little effect until recently? Our findings suggest that removal restrictions were redundant as long as there were political norms against partisan involvement in independent agencies and political agreement on the missions of independent agencies. In other words, bipartisan norms pushed presidents to leave independent agencies alone, and presidents generally agreed with the agencies’ actions anyway. The statutory mechanisms protecting agency heads thus served little additional purpose.
But recently, these political norms have been breaking down, and the overlap of Republicans’ and Democrats’ preferences regarding the work of independent agencies has increasingly thinned. This phenomenon holds true across many independent agencies—even those originally created with near-unanimous bipartisan support, such as the Federal Reserve. So, although the statutory protections for the Federal Reserve have likely been duplicative of other circumstances promoting its independence, increasing division over the agency’s independence and its proper role in the modern economy suggests that the question whether its removal protections are constitutional could not be more important.
On the constitutional issue, our research might lend support to both sides on the Court. Some justices might view our findings as evidence that removal restrictions meaningfully interfere with the president’s ability to execute laws in a way that aligns with the will of the electorate. They might worry that independent agencies form a fourth branch of government, unaccountable to the president or any elected official, and thus unaccountable to the American people. If today’s political climate leads removal restrictions to frustrate the president’s exercise of his constitutional duties, then now is precisely the right time to set them aside.
Other justices might see the same findings as evidence that removal restrictions protect the law Congress enacted and ensure the president doesn’t interfere with the faithful execution of that law. For these justices, presidential interference taints agency expertise and independent judgment with base political and electoral interests. If today’s political climate leaves these statutory protections as the last line of defense, then it is more important than ever to stand by and enforce them against the president.
Regardless of which argument one thinks is most legally or politically compelling, both sides at the Supreme Court (and anyone else in the debate) should recognize the irony embedded in our findings. The Supreme Court might well rule removal restrictions to be unconstitutional, only and precisely when they have started to matter. Conversely, upholding these restrictions against attack will put more pressure on them than they have historically borne.
This article was adapted and revised with permission from an earlier essay by two of the authors, originally published with Bloomberg Law.
Authors’ Disclosures: The authors report no conflicts of interest. You can read our disclosure policy here.
Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.
Subscribe here for ProMarket’s weekly newsletter, Special Interest, to stay up to date on ProMarket’s coverage of the political economy and other content from the Stigler Center.
